In the context of Elementary's anomaly detection tests,

detected_at

typically refers to the timestamp when a particular test detects an anomaly. This is the moment the system identifies that data is outside the expected range based on historical patterns and thresholds set for alerting.

On the other hand,

generated_at

usually denotes the timestamp when the data itself was generated or created. In the case of monitoring event data freshness, for example,

generated_at

could represent when an event actually occurred or when it was first recorded in the database.

Elementary's anomaly detection tests can monitor the freshness of data by looking at the expected time it takes for an event to be loaded into the database, which involves comparing an

event_timestamp_column

(when the event occurred) with an

update_timestamp_column

(when it was loaded into the database) . This is an example where understanding the difference between when data was generated and when it was detected (or loaded) is crucial for accurate monitoring.

However, based on the information provided in the cited documents, there are no explicit definitions for

detected_at

or

generated_at

. The terms seem to be more general database or data warehousing concepts rather than specific configurations or parameters used by Elementary. Therefore, for detailed technical specifics on how these terms are implemented or used within Elementary, I wasn't able to find any information about your question.